Back to portfolio

CASE 02 / Material traceability

Material traceability on demand

I developed a VBA automation that reads TXT and web sources, structures supply and material data, and connects it to the order book for on-demand updates.

02.1 / Context

Consolidating the information took 3–4 hours.

The team needed to consolidate supply, material characteristics, movements, orders and forecasts. Execution was long, depended on one person and was concentrated in one shift.

Investigating the problem

The routine gathered supply information, material characteristics, movements, the order book and forecasts. Effort was concentrated in extracting and connecting these inputs before a usable view could be produced.

Implementation constraints

Consolidation depended on one person and one shift. The solution needed to use available resources and make updates accessible when operations required them.

TimeThe routine took 3–4 hours per execution.
DependencyOperational knowledge was concentrated around one person.
FrequencyPreparation cost limited updates throughout the day.

02.2 / Development

Data extraction, processing and integration

The solution read TXT files, performed web extraction, split content into arrays, applied structuring rules and connected sources to the order book. Cognos information was included without exposing internal queries.

TXT + WEB: Supply and raw material sources → VBA: Repeatable extraction → Structure: Split · arrays · processing rules → Integration: Connection to the order book and supply context → Operations: On-demand updates · different shifts01TXT + WEBSupply and rawmaterial sources02VBARepeatableextraction03StructureSplit · arrays ·processing rules04IntegrationConnection to theorder book andsupply context05OperationsOn-demand updates· differentshifts
  1. TXT + WEB

    Supply and raw material sources

  2. VBA

    Repeatable extraction

  3. Structure

    Split · arrays · processing rules

  4. Integration

    Connection to the order book and supply context

  5. Operations

    On-demand updates · different shifts

Select a stage to explore its role in the flow.

← Scroll horizontally to explore the diagram →

Conceptual representation · no internal dataVBA · TXT · web extraction · IBM Cognos

02.3 / Project decisions

VBA brought the automation closer to the process.

The architecture prioritised maintenance, availability and cross-shift use.

Technical risk

A solution outside the available environment can make adoption harder.

Use a tool embedded in the process

VBA was already available in the operating environment and avoided introducing a separate platform.

Technical risk

Combining extraction and association rules makes changes harder to locate.

Separate the stages

Extraction, transformation, association and output were structured for maintenance and verification.

Technical risk

A fast routine remains limited if only one person can run it.

Enable on-demand use

Execution no longer depended on a specific time and became available across shifts.

From decision to implementation

  1. Extract

    The routine reads TXT files and web sources, incorporating information obtained through Cognos.

  2. Structure and connect

    Content is split into arrays, processed with rules and associated with the order book and supply context. Each stage has a distinct responsibility.

  3. Update on demand

    Execution became available to different people and shifts, including critical closing periods, reducing dependence on manual consolidation.

3–4 h → <1 minper update

02.4 / Evidence and outcomes

Updates began to match the pace of operations.

Updates became more frequent, available to different people and accessible during critical month-end periods.

  • less time spent on manual consolidation;
  • updates available across shifts;
  • less individual dependency;
  • lower pressure during month-end closing.

Verification / interpreting the evidence

Time per run

The routine went from 3–4 hours to under one minute per update. The comparison describes this routine, not the duration of the entire industrial process.

Engineering criterion

Speed does not replace reliability. Separating extraction, processing and association makes it clear where changes in sources or rules can affect the output.

Limits of the public evidence

There is no public timing series or integrity test log. The case describes the documented architecture without claiming an error rate or unsubstantiated controls.

02.5 / Synthesis

Learning

Separating extraction, processing and integration supported maintenance of a routine needed across different shifts.